\n \n \n
\n
\n\n \n \n \n \n \n Bibliometric-enhanced information retrieval: 13th international BIR workshop (BIR 2023).\n \n \n \n\n\n \n Frommholz, I.; Mayr, P.; Cabanac, G.; and Verberne, S.\n\n\n \n\n\n\n In Kamps, J.; Goeuriot, L.; Crestani, F.; Maistro, M.; Joho, H.; Davis, B.; Gurrin, C.; Kruschwitz, U.; and Caputo, A., editor(s),
Advances in information retrieval, pages 392–397, Cham, 2023. Springer Nature Switzerland\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{frommholz_bibliometric-enhanced_2023,\n\taddress = {Cham},\n\ttitle = {Bibliometric-enhanced information retrieval: 13th international {BIR} workshop ({BIR} 2023)},\n\tcopyright = {All rights reserved},\n\tisbn = {978-3-031-28241-6},\n\tabstract = {The 13th iterationof the Bibliometric-enhanced Information Retrieval (BIR) workshop series will take place at ECIR 2023 as a full-day workshop. BIR tackles issues related to, for instance, academic search and recommendation, at the intersection of Information Retrieval, Natural Language Processing, and Bibliometrics. As an interdisciplinary scientific event, BIR brings together researchers and practitioners from the Scientometrics/Bibliometrics community on the one hand, and the Information Retrieval community on the other hand. BIR is an ever-growing topic investigated by both academia and the industry.},\n\tbooktitle = {Advances in information retrieval},\n\tpublisher = {Springer Nature Switzerland},\n\tauthor = {Frommholz, Ingo and Mayr, Philipp and Cabanac, Guillaume and Verberne, Suzan},\n\teditor = {Kamps, Jaap and Goeuriot, Lorraine and Crestani, Fabio and Maistro, Maria and Joho, Hideo and Davis, Brian and Gurrin, Cathal and Kruschwitz, Udo and Caputo, Annalina},\n\tyear = {2023},\n\tpages = {392--397},\n}\n\n
\n
\n\n\n
\n The 13th iterationof the Bibliometric-enhanced Information Retrieval (BIR) workshop series will take place at ECIR 2023 as a full-day workshop. BIR tackles issues related to, for instance, academic search and recommendation, at the intersection of Information Retrieval, Natural Language Processing, and Bibliometrics. As an interdisciplinary scientific event, BIR brings together researchers and practitioners from the Scientometrics/Bibliometrics community on the one hand, and the Information Retrieval community on the other hand. BIR is an ever-growing topic investigated by both academia and the industry.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Parallelization Strategies for Graph-Code-Based Similarity Search.\n \n \n \n \n\n\n \n Steinert, P.; Wagenpfeil, S.; Mc Kevitt, P.; Frommholz, I.; and Hemmje, M.\n\n\n \n\n\n\n
Big Data and Cognitive Computing, 7(2): 70. April 2023.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@article{steinert_parallelization_2023,\n\ttitle = {Parallelization {Strategies} for {Graph}-{Code}-{Based} {Similarity} {Search}},\n\tvolume = {7},\n\tcopyright = {All rights reserved},\n\tissn = {2504-2289},\n\turl = {https://www.mdpi.com/2504-2289/7/2/70},\n\tdoi = {10.3390/bdcc7020070},\n\tabstract = {The volume of multimedia assets in collections is growing exponentially, and the retrieval of information is becoming more complex. The indexing and retrieval of multimedia content is generally implemented by employing feature graphs. Feature graphs contain semantic information on multimedia assets. Machine learning can produce detailed semantic information on multimedia assets, reflected in a high volume of nodes and edges in the feature graphs. While increasing the effectiveness of the information retrieval results, the high level of detail and also the growing collections increase the processing time. Addressing this problem, Multimedia Feature Graphs (MMFGs) and Graph Codes (GCs) have been proven to be fast and effective structures for information retrieval. However, the huge volume of data requires more processing time. As Graph Code algorithms were designed to be parallelizable, different paths of parallelization can be employed to prove or evaluate the scalability options of Graph Code processing. These include horizontal and vertical scaling with the use of Graphic Processing Units (GPUs), Multicore Central Processing Units (CPUs), and distributed computing. In this paper, we show how different parallelization strategies based on Graph Codes can be combined to provide a significant improvement in efficiency. Our modeling work shows excellent scalability with a theoretical speedup of 16,711 on a top-of-the-line Nvidia H100 GPU with 16,896 cores. Our experiments with a mediocre GPU show that a speedup of 225 can be achieved and give credence to the theoretical speedup. Thus, Graph Codes provide fast and effective multimedia indexing and retrieval, even in billion-scale use cases.},\n\tlanguage = {en},\n\tnumber = {2},\n\turldate = {2023-06-06},\n\tjournal = {Big Data and Cognitive Computing},\n\tauthor = {Steinert, Patrick and Wagenpfeil, Stefan and Mc Kevitt, Paul and Frommholz, Ingo and Hemmje, Matthias},\n\tmonth = apr,\n\tyear = {2023},\n\tpages = {70},\n}\n\n
\n
\n\n\n
\n The volume of multimedia assets in collections is growing exponentially, and the retrieval of information is becoming more complex. The indexing and retrieval of multimedia content is generally implemented by employing feature graphs. Feature graphs contain semantic information on multimedia assets. Machine learning can produce detailed semantic information on multimedia assets, reflected in a high volume of nodes and edges in the feature graphs. While increasing the effectiveness of the information retrieval results, the high level of detail and also the growing collections increase the processing time. Addressing this problem, Multimedia Feature Graphs (MMFGs) and Graph Codes (GCs) have been proven to be fast and effective structures for information retrieval. However, the huge volume of data requires more processing time. As Graph Code algorithms were designed to be parallelizable, different paths of parallelization can be employed to prove or evaluate the scalability options of Graph Code processing. These include horizontal and vertical scaling with the use of Graphic Processing Units (GPUs), Multicore Central Processing Units (CPUs), and distributed computing. In this paper, we show how different parallelization strategies based on Graph Codes can be combined to provide a significant improvement in efficiency. Our modeling work shows excellent scalability with a theoretical speedup of 16,711 on a top-of-the-line Nvidia H100 GPU with 16,896 cores. Our experiments with a mediocre GPU show that a speedup of 225 can be achieved and give credence to the theoretical speedup. Thus, Graph Codes provide fast and effective multimedia indexing and retrieval, even in billion-scale use cases.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n An Extended Study of Search User Interface Design Focused on Hofstede's Cultural Dimensions.\n \n \n \n \n\n\n \n Chessum, K.; Liu, H.; and Frommholz, I.\n\n\n \n\n\n\n In Holzinger, A.; Da Silva, H. P.; Vanderdonckt, J.; and Constantine, L., editor(s),
Computer-Human Interaction Research and Applications, volume 1882, pages 130–152, Cham, 2023. Springer Nature Switzerland\n
Series Title: Communications in Computer and Information Science\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 2 downloads\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{holzinger_extended_2023,\n\taddress = {Cham},\n\ttitle = {An {Extended} {Study} of {Search} {User} {Interface} {Design} {Focused} on {Hofstede}'s {Cultural} {Dimensions}},\n\tvolume = {1882},\n\tcopyright = {All rights reserved},\n\tisbn = {978-3-031-41961-4 978-3-031-41962-1},\n\turl = {https://link.springer.com/10.1007/978-3-031-41962-1_7},\n\tabstract = {Geert Hofstede’s classic cultural model has been studied and applied to website design for a number of years. In this paper we examine if Geert Hofstede’s six cultural dimensions can also be applied to search user interface design. Two user studies have been conducted to evaluate the culturally designed search user interfaces, and the findings are reported in this paper. Our first study comprised of 148 participants from different cultural backgrounds. The second study was smaller with 25 participants, also from different cultural backgrounds. The results from these studies have been analyzed to ascertain if Hofstede’s cultural dimensions are suitable for understanding users’ preferences for search user interface design. Whilst the key findings from these studies suggest Hofstede cross-cultural dimensions can be used to model users’ preferences on search interface design, further work is still needed for particular cultural dimensions to reinforce the conclusions.},\n\tlanguage = {en},\n\turldate = {2023-08-30},\n\tbooktitle = {Computer-{Human} {Interaction} {Research} and {Applications}},\n\tpublisher = {Springer Nature Switzerland},\n\tauthor = {Chessum, Karen and Liu, Haiming and Frommholz, Ingo},\n\teditor = {Holzinger, Andreas and Da Silva, Hugo Plácido and Vanderdonckt, Jean and Constantine, Larry},\n\tyear = {2023},\n\tdoi = {10.1007/978-3-031-41962-1_7},\n\tnote = {Series Title: Communications in Computer and Information Science},\n\tpages = {130--152},\n}\n\n
\n
\n\n\n
\n Geert Hofstede’s classic cultural model has been studied and applied to website design for a number of years. In this paper we examine if Geert Hofstede’s six cultural dimensions can also be applied to search user interface design. Two user studies have been conducted to evaluate the culturally designed search user interfaces, and the findings are reported in this paper. Our first study comprised of 148 participants from different cultural backgrounds. The second study was smaller with 25 participants, also from different cultural backgrounds. The results from these studies have been analyzed to ascertain if Hofstede’s cultural dimensions are suitable for understanding users’ preferences for search user interface design. Whilst the key findings from these studies suggest Hofstede cross-cultural dimensions can be used to model users’ preferences on search interface design, further work is still needed for particular cultural dimensions to reinforce the conclusions.\n
\n\n\n
\n\n\n
\n\n\n
\n\n\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Who can Submit an Excellent Review for this Manuscript in the Next 30 Days? - Peer Reviewing in the Age of Overload.\n \n \n \n \n\n\n \n Alhoori, H.; Fox, E. A.; Frommholz, I.; Liu, H.; Coupette, C.; Rieck, B. A.; Ghosal, T.; and Wu, J.\n\n\n \n\n\n\n In
2023 ACM/IEEE Joint Conference on Digital Libraries (JCDL), pages 319–320, Santa Fe, NM, USA, June 2023. IEEE\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{alhoori_who_2023,\n\taddress = {Santa Fe, NM, USA},\n\ttitle = {Who can {Submit} an {Excellent} {Review} for this {Manuscript} in the {Next} 30 {Days}? - {Peer} {Reviewing} in the {Age} of {Overload}},\n\tcopyright = {All rights reserved},\n\tisbn = {9798350399318},\n\tshorttitle = {Who can {Submit} an {Excellent} {Review} for this {Manuscript} in the {Next} 30 {Days}?},\n\turl = {https://ieeexplore.ieee.org/document/10265918/},\n\tdoi = {10.1109/JCDL57899.2023.00077},\n\tabstract = {With millions of research articles published yearly, the peer review process is in danger of collapsing, especially in “hot” areas with popular conferences. Challenges arise from the large number of manuscripts submitted, skyrocketing use of preprint archives and institutional repositories, problems regarding the identification and availability of experts, conflicts of interest, and bias in reviewing. Such issues can affect the integrity of the reviewing process as well as the timeliness, quality, credibility, and reproducibility of research articles. Several solutions and systems have been suggested, but none work well, and neither authors nor editors are happy with how long it takes to complete reviewing the submitted research. This panel addresses these challenges and potential solutions, including digital libraries that recommend reviewers, as well as broader issues like opportunities for identifying peer reviewers for scholarly journals by engaging doctoral students and postdocs, as well as those who recently completed their Ph.D.},\n\tlanguage = {en},\n\turldate = {2023-10-04},\n\tbooktitle = {2023 {ACM}/{IEEE} {Joint} {Conference} on {Digital} {Libraries} ({JCDL})},\n\tpublisher = {IEEE},\n\tauthor = {Alhoori, Hamed and Fox, Edward A. and Frommholz, Ingo and Liu, Haiming and Coupette, Corinna and Rieck, Bastian A. and Ghosal, Tirthankar and Wu, Jian},\n\tmonth = jun,\n\tyear = {2023},\n\tpages = {319--320},\n}\n\n
\n
\n\n\n
\n With millions of research articles published yearly, the peer review process is in danger of collapsing, especially in “hot” areas with popular conferences. Challenges arise from the large number of manuscripts submitted, skyrocketing use of preprint archives and institutional repositories, problems regarding the identification and availability of experts, conflicts of interest, and bias in reviewing. Such issues can affect the integrity of the reviewing process as well as the timeliness, quality, credibility, and reproducibility of research articles. Several solutions and systems have been suggested, but none work well, and neither authors nor editors are happy with how long it takes to complete reviewing the submitted research. This panel addresses these challenges and potential solutions, including digital libraries that recommend reviewers, as well as broader issues like opportunities for identifying peer reviewers for scholarly journals by engaging doctoral students and postdocs, as well as those who recently completed their Ph.D.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n What if ChatGPT wrote the Abstract? - Explainable Multi-Authorship Attribution with a Data Augmentation Strategy.\n \n \n \n \n\n\n \n Silva, K.; and Frommholz, I.\n\n\n \n\n\n\n In
Proceedings of the IACT - The 1st International Workshop on Implicit Author Characterization from Texts for Search and Retrieval (IACT@SIGIR 2023), pages 38–48, Taipei, Taiwan, 2023. CEUR-WS.org\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{silva_what_2023,\n\taddress = {Taipei, Taiwan},\n\ttitle = {What if {ChatGPT} wrote the {Abstract}? - {Explainable} {Multi}-{Authorship} {Attribution} with a {Data} {Augmentation} {Strategy}},\n\tcopyright = {Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Licence (CC-BY-NC-ND)},\n\turl = {https://ceur-ws.org/Vol-3477/paper6.pdf},\n\tabstract = {Active discussions have been conducted regarding implications and issues associated with Large Language Models (LLMs) such as ChatGPT across various domains. One particular concern is the effect of machinegenerated texts, which include a new category in authorship attribution models: machine-generated text resembling human text in topic and writing style. Differentiating human-vs-AI-written text in scientific articles is crucial for several reasons. In this work, we approach this issue from a multi-authorship perspective by investigating automatically generated abstracts. We propose a multimodal transformer which combines handcrafted stylometric features with deep learning-based text features to perform multi-author attribution. We demonstrate the effectiveness of this approach on a curated dataset of 1000 samples and discuss its explainability via the Local Interpretable Model-agnostic Explanations (LIME) framework.},\n\tlanguage = {en},\n\tbooktitle = {Proceedings of the {IACT} - {The} 1st {International} {Workshop} on {Implicit} {Author} {Characterization} from {Texts} for {Search} and {Retrieval} ({IACT}@{SIGIR} 2023)},\n\tpublisher = {CEUR-WS.org},\n\tauthor = {Silva, Kanishka and Frommholz, Ingo},\n\tyear = {2023},\n\tpages = {38--48},\n}\n\n
\n
\n\n\n
\n Active discussions have been conducted regarding implications and issues associated with Large Language Models (LLMs) such as ChatGPT across various domains. One particular concern is the effect of machinegenerated texts, which include a new category in authorship attribution models: machine-generated text resembling human text in topic and writing style. Differentiating human-vs-AI-written text in scientific articles is crucial for several reasons. In this work, we approach this issue from a multi-authorship perspective by investigating automatically generated abstracts. We propose a multimodal transformer which combines handcrafted stylometric features with deep learning-based text features to perform multi-author attribution. We demonstrate the effectiveness of this approach on a curated dataset of 1000 samples and discuss its explainability via the Local Interpretable Model-agnostic Explanations (LIME) framework.\n
\n\n\n
\n\n\n\n\n\n